Stijn De Haes

I have left the node running so I can debug it further if needed :)

I still see this issue happening @ctalledo. A `sudo systemctl start sysbox-mgr` fixes the problem.

I would love to use this for enabling bin packing like explained here: https://kubernetes.io/docs/concepts/scheduling-eviction/resource-bin-packing/

I would be fine with having a setting like GKE has, this would solve my use case. It probably does not solve every use case out there, but I can...

Hi I am not sure why every numbers is infinity. I updated my example project: https://github.com/stijndehaes/play-prometheus-filters-example-app to use the latest version but could not reproduce your error. Can you give...

@cartermckinnon suggested it might have something to do with EBS lazy block fetching resulting in slow startup for the executables we need: https://github.com/awslabs/amazon-eks-ami/issues/1696#issuecomment-2037546262

One way we can speed things up is to make sure that kubelet and containerd are loaded earlier in the chain. We could for example: - have containerd and kubelet...

@cartermckinnon I am willing to give it a try. As to your launch steps question, I am not enterily sure I understand what you want. Can you elaborate?

Here is the new result, with fast restore on the snapshot enabled. Sadly I see roughly the same timing, I also see similar delays in the logs.  ``` sh-5.2#...

@dims Sorry it took so long to get back to you. I have restried with fast restore and got some better results. I takes around 50s with fast restore (still...